Machine learning

Introduction to neural network technologies
  1. Machine learning as function approximation; statistical and neural motivations
  2. Comments on overfitting: accidentally peeking at data (instead look for general ways to prevent overfitting), cross-validation
  3. Backpropagation and the chain rule + other optimization algorithms (Nesterov, Adam etc.)
  4. Computer vision: convolutional neural networks
  5. Computer vision: image segmentation
  6. Computer vision: landmark detection
  7. Stream of data: RNNs
  8. Natural language processing
  9. Generative neural networks: introduction
  10. Generative neural networks: upsampling
  11. Generative neural networks: overview of architectures
  12. Generative neural networks: deepfakes
  13. Generative neural networks: style transfer
  14. Generative neural networks: image relighting, changing specific attributes [1]
  15. Generative neural networks: differentiable physics
  16. Resnet etc.
  17. Reinforcement learning
  18. Transfer learning: transfer, progressive neural networks etc.
Projects
  1. Random walk of character shapes
  2. Simulating randomness
  3. AI colorize B/W image
  4. AI image beautification
  5. Create characters that look like characters but not existing ones
  6. Peeking into neural networks -- generative networks?
  7. Neural style transfer on tunes

https://openai.com/blog/generative-models/
https://towardsdatascience.com/gan-by-example-using-keras-on-tensorflow-backend-1a6d515a60d0

https://towardsdatascience.com/@joseph.rocca

https://nptel.ac.in/courses/128/106/128106011/

No comments:

Post a Comment